[Mimedefang] MIMEDefang 2.57-BETA-1 is available

[David F. Skoll]

Steffen Kaiser wrote:

> Would it possible to think about to _save_ the state between all the
> functions?

It's complex and I think has too much overhead.  In addition, pretty much
all the interesting information available at earlier stages is *also*
made available to later stages, so there's usually no need to save state
between callbacks.  (In fact, I have never encountered a situation in
which it's necessary.)

> The man page itself tells what to do, however, when the SMTP dialogue is
> rather straigth forward, I mean, when the client does not pause before
> sending the next data, the probability is high, that the same slave is
> still idle and can be passed to request to.

:-)  I prefer to program a deterministic machine rather than a probabalistic
one.

[...]

> I wonder if there will be any benefit at all.

I do not think so, because as I said, all the information from earlier
phases is already available in later phases.  In the rare case in which you
do an expensive computation in filter_sender, for example, and you need the
results later on in filter_begin, you can just save to a file.  I can
definitely see a use for library functions to save/restore a hash full
of state.

Regards,

David.